The dorsal cochlear nucleus (DCN) is a laminar structure with intricate internal organization based on isofrequency sheets generated by auditory nerve (AN) projections to the DCN. A second organization is provided by axons of cochlear nucleus granule cells, which run orthogonal to the isofrequency sheets. The DCN contains several types of inhibitory interneurons which play an important role in the responses to sound of its principal cells. The goals of this project are to provide new information about the functional organization of the DCN, to develop new methods for characterizing neurons with complex responses to sound like those of DCN, and to pursue a hypothesis about the functional role of the DCN in hearing. The first and second aims are designed to characterize the effects of excitatory and inhibitory inputs to DCN principal cells in shaping their response properties. In addition to AN fibers, the DCN receives tonotopically organized excitatory and inhibitory inputs from the posteroventral cochlear nucleus (PVCN). In the first aim, excitotoxic lesions of the PVCN will be made using kainic acid or ibotenic acid to eliminate the PVCN cellular input to DCN, while sparing the AN inputs; the effects of the PVCN inputs will be inferred by studying the DCN before and after such lesions. The second aim will use cross-correlation of spike trains recorded simultaneously from pairs of units to search for as-yet-unrecognized inhibitory inputs to DCN principal cells. The presence of such inputs is suggested by experiments currently underway. The third aim will apply a new non-linear characterization method to the problem of describing the input/output relationships of DCN cells. Current methods for characterizing auditory neurons fail to predict the responses of complex neurons, such as DCN principal cells, to biologically important stimuli. Success in this aim could have wide application to the study of other parts of the central auditory system. The fourth aim will study the representation, in DCN and lateral superior olive (LSO), of spectrally-encoded sound localization cues produced by the pinna. Current experiments suggest that the DCN is specialized to extract and encode the information-bearing features of such stimuli; the LSO is another part of the central auditory system that is likely to be important in analysis of these features. A goal of this portion of the project is to compare the representation of spectral sound localization features in DCN and LSO, which are parallel pathways in the brainstem auditory system, so that hypotheses about the processing of these features can be refined. The DCN is a valuable model for neuroscience research because it is a complex neural machine which shares some features with cerebellar and cerebral cortex, but is still close enough to the periphery that its inputs are easy to characterize and manipulate. It is interesting to clinical otolaryngology because it is the most likely site, for technical reasons, for a human cochlear nucleus prosthesis.